Genre Classification of Literary Texts through Deep Learning Methods (Based on the Russian-language Fanfiction Electronic Database)

The paper is devoted to the genre classification of literary texts using artificial intelligence methods. The research material is fanfiction texts included in the supplemented Russian-language electronic fanfiction database, the data source for which was the popular specialized resource "Kniga Fanfikov". The main purpose of the study is to develop a version of a neural network model for classifying literary texts by genre based on fanfiction works. The work also examines the content and lexical and statistical features of the fanfiction genres. The results obtained include three genre classifiers of different types fine-tuned on fanfiction data and tested on the material of belletristic literature. The results of the research can be applied both to solve practical problems (e.g. the classification of literary works from electronic libraries, databases), and in scientific research in the field of computational linguistics and digital humanities.